[Bug 72210] [Regression BYT] S3 response time large than before

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=72210

--- Comment #10 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to comment #9)
> (In reply to comment #7)
> When you say "between commits", do you mean between the commit
> 4aeebd7443e36b0a40032e518a9338f48bd27efc and the latest commit on
> drm-intel-nightly ?

Roughly. I just wanted to measure 4aeebd7443e36b0a40032e518a9338f48bd27efc and
its parent (so that we can correlate how the change in xrandr corresponds to
the initcall times) and then confirm the problem still exists unadulterated on
drm-intel-nightly.

> And there exits two bugs on BYT DP port(Bug 72896 and Bug 72897), the screen
> will become black after load I915 driver. So I want to know whether it's OK
> to test with eDP attached?

The issue is that eDP caches its EDID upon initialisation so we cannot use
xrandr to time how long it takes to read the EDID over the aux channel, and
therefore we cannot use this method to measure the impact of the IRQ paths.

To use eDP for this testing, you would need to also apply:

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 885d271942b3..ab339cc63640 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3586,6 +3586,7 @@ static bool intel_edp_init_connector(struct intel_dp
*intel_dp,
        /* We now know it's not a ghost, init power sequence regs. */
        intel_dp_init_panel_power_sequencer_registers(dev, intel_dp,
power_seq);

+#if 0
        edid = drm_get_edid(connector, &intel_dp->adapter);
        if (edid) {
                if (drm_add_edid_modes(connector, edid)) {
@@ -3599,6 +3600,9 @@ static bool intel_edp_init_connector(struct intel_dp
*intel_dp,
        } else {
                edid = ERR_PTR(-ENOENT);
        }
+#else
+       edid = ERR_PTR(-ENOENT);
+#endif
        intel_connector->edid = edid;

        /* prefer fixed mode from EDID if available */

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20140120/98457820/attachment.html>